Hacker News Reader

Comment by JTenerife | Hacker News Reader

Younger people are not smarter. They might learn faster new things. On the other hand, older people understand related new things much better as they already have a large context (experience). The biggest difference of my current self (44) to my younger self is that I did spend much more energy in my projects when I was younger. I created results much faster at the cost of limited consideration. Now, I can still burn for a while, but not as long as when I was younger.

View on HN

I used to be able to burn longer when I was twenty, but I'd spend that fuel on shit I've learned doesn't matter like 100% test case coverage, setting up continuous integration servers, and load testing against massive amounts of traffic I'd never see.

I could go on and on about the stuff I'd do I don't do anymore.

It's not that those things don't have their place, or that all young programmers are guilty of premature optimizations, but there's no difference in terms of productivity between doing nothing and doing things that don't matter at all. It took me a while to learn that, and I still catch myself wasting time.

Do you ever get asked about this? e.g. what you don't do anymore as a result of learning from experience

Either in your current role / in interviews / on the street?

not that i can recall. why do you ask?

It seems like you have learned a tremendous amount throughout your career that others could benefit from, that's why! Not learning from you flies in the face of wisdom (1)(2).

I'm not afraid to generalize this either. I see so many people and organizations repeating the same mistakes, of their own, and of others', over and over. I envision an organization built around the idea of learning from others' experience. It's a near mythical creature, but it could run circles around its peers, who would doggedly pursue finding out its "secret" and promptly discounting it when they hear "learn from others' experience / history."

This turned into a mini-rant, but it's because I am incredulous that you and others like you have learned so much yet the potential of that knowledge is so rarely tapped into.

(1)“Fools learn from experience. I prefer to learn from the experience of others.” ― Otto von Bismarck (2) Why Don't We Learn from History by B.H. Liddell Hart https://www.amazon.com/Why-Dont-We-Learn-History/dp/09850811...

I find that - very broadly speaking - old guns are better at strategy and young guns are better at tactics. The energy and interest of the young lead them to find the local maxima more readily, and the experience of the old help you find the real maxima rather than the local. Or put another way, young'uns will get you out of the rabbit-hole more quickly, but old'uns will take you down fewer rabbit-holes. Both mindsets work well together.

YMMV, caveat, caveat, caveat, etc, roughly speaking, &c

Would you mind expanding a little bit more, from your perspective, on what falls into tactics and what falls into strategy?

Generally what marklgr said. By tactics, I mean being up to date with the latest lib for -foo- and being on top of what the cool kids are doing, chasing down the rabbit-hole to find that one setting to improve db performance. Individual, almost atomic tasks, especially those in a changing landscape.

Strategy is more about planning, forethought, and knowing ahead of time which avenues are worth pursuing and which aren't. As a very simplistic example, a tactician might spend a day down a rabbit-hole and improve a db's performance by a small but significant amount, and a strategist would not have spent that day, knowing that that db is going to be turned off in two weeks. Not the best example, really (and it implies that you're only one or the other, which isn't true at all). There is also overlap between the two.

Perhaps a military analogue - tactics is about how to take the forts (shorter timeframes, clearer objectives, obvious goals), and strategy is about determining which forts are worth taking (long-term timeframes, murkier objectives, sometimes unclear goals, concern for secondary effects). The difference is also in scope, I guess. Being a good tactician benefits from energy and focused interest, something which the young'uns tend to have more of (caveat, caveat), and strategy benefits from forethought and experience, something which the old'uns tend to have more of (caveat, caveat, &c)

In that context, to me, tactics is writing & debugging code, compiling stuff, settings things up to work with other libs--ie. mostly hands-on tasks.

Strategy is more about planning, architecture, knowing about the real pros and cons of different solutions eg. libs, tools or languages.

At 37 I'm much faster at debugging today than I was at 27. And 10 years ago, I was consistently one of the fastest debuggers in a company of 15-20 programmers. Debugging's always been one of my strongest programming skills.

So debugging, no, definitely not. I can spot root-causes of bugs now even easier than I used to be able to.

And one of the big new skills is that I'm much more able to 'guess' where a bug is in someone else's totally new code-base than I previously was able to.

In fact, I also write code faster as I get the general gist of it done much faster first time.

Some people specialize, and get more and more efficient at the same specific tasks; but I believe it's more common to widen one's experience in various areas (languages, tools, platforms, non-functional requirements etc.), leading to some lack of practice in the hands-on tasks one used to do everyday in the past, and also making it more difficult to commit everything to memory. On the other hand, one generally gets a much broader picture.

For instance, I used to be able to perftune UNIX and DB (mostly Oracle on Solaris) pretty well, knowing various kernel parameters by heart and all the tools of the day (tkprof, Cockroft and co). Now I guess I could still get by after some brush-up, but I'm not anywhere near as efficient as I used to be on that specific task. And I still write code, but I don't know every single method of every API anymore. But now, I can design full solutions, from choosing the hardware to setting up platform, languages, monitoring, high-availability, backups, security etc. Not because I'm smarter, but because I've been exposed to all that along the years.

Breadth of experience often turns into strategy. Lack of experience tends to show up as when you only have a hammer, every problem looks like a nail. At this point in my career I've written production code in many languages/frameworks and was even a DBA at one point. I've also worked in many types of businesses. Now I can look at a given problem and provide a range of possible solutions. Sometimes those solutions do not involve writing any code at all.

I've seen several examples where the younger developer (often me) comes up with some really clever code to make a certain thing run really fast, and the older developer coming along and pointing out that my underlying premise is wrong.

I agree with all your sentiments. I think the key thing is that younger people tend to equate efficiency with lines of code written. As you get older, you realize that pausing long enough to consider all implications leads to better solutions. Fast and hacky is fine for prototypes but (hopefully) eventually it needs to scale and be stable and supportable.

I also think that the Zuckerberg quote also points to the startup mentality. They want 22 year olds because you can get them to work 100 hour weeks against the promise of an IPO, not because they are "smarter". Hard to convince a 40 year old with two kids to do the same. There has already been a shift to more traditional CS companies (IBM, etc.) as the lack of IPOs has soured some people on the startup dream of being employee #3 at the next Uber of XXX.

Agreed. My current self can understand complex concepts faster than my younger self. Understanding the innards of a distributed database is easier now than ten years back.

> Younger people are not smarter. They might learn faster new things.

I was a young person once (no, seriously!). I _thought_ I was learning things fast back then. When I got older, I realized I was trading speed for depth.

Young people:

    1. Learn faster (better memory)

    2. Can keep their attention focused longer.

Number 1 actually becomes a bane with old people: my relatives over 70 actively developed an active reticence to learn anything new: I even theorized that they are instinctively protecting the limited amount of functional short term memory for vital tasks.

    > Learn faster (better memory)

And yet, when I needed to learn XSLT (and then React), I was able to simply inhale it in one go because of my experience of similar technologies and functional programming, which come about from having been doing this quite a while.

Younger and quicker developers took a lot longer over it and to find their feet because there were many new concepts there for them.

As you get older, there's less to you haven't already learned.

Let's put aside the unsubstantiated claims about biology (and old doesn't mean 70 yrs in this context). Jobs that rely heavily on attention span and learning speed can also quickly get boring and unfulfilling--the more experienced folks probably know to steer away from that.

Reply on news.ycombinator.com